Photocatalytic degradation of eleven microcystin analogues andnodularin by TiO2 coated glass microspheres

Pestana, C. J., Edwards, C., Prabhu, R., Robertson, P. K. J., & Lawton, L. A. (2015). Photocatalytic degradationof eleven microcystin analogues and nodularin by TiO2 coated glass microspheres. Journal of HazardousMaterials, 300, 347-353. https://doi.org/10.1016/j.jhazmat.2015.07.016

Published in:Journal of Hazardous Materials

Document Version:Peer reviewed version

Queen's University Belfast - Research Portal:Link to publication record in Queen's University Belfast Research Portal

Publisher rightsCopyright © 2015 Elsevier B.V. All rights reserved.This manuscript version is made available under the CC-BY-NC-ND 4.0 license http://creativecommons.org/licenses/by-nc-nd/4.0/), whichpermits distribution and reproduction for non-commercial purposes, provided the author and source are cited.

General rightsCopyright for the publications made accessible via the Queen's University Belfast Research Portal is retained by the author(s) and / or othercopyright owners and it is a condition of accessing these publications that users recognise and abide by the legal requirements associatedwith these rights.

Take down policyThe Research Portal is Queen's institutional repository that provides access to Queen's research output. Every effort has been made toensure that content in the Research Portal does not infringe any person's rights, or applicable UK laws. If you discover content in theResearch Portal that you believe breaches copyright or violates any law, please contact openaccess@qub.ac.uk.

Download date:30. May. 2021

1

Photocatalytic degradation of eleven microcystin variants and

nodularin by TiO2 coated glass microspheres

Carlos J. Pestanaa, Christine Edwardsa, Radhakrishna Prabhua, Peter K.J.

Robertsonb*, Linda A. Lawtona.

aInnovation, Design and Sustainability (IDeaS) Research Institute, Robert Gordon

University, Riverside East, Garthdee Road, Aberdeen AB10 7GJ, UK

bCentre for the Theory and Application of Catalysis (CenTACat), School of Chemistry

and Chemical Engineering, Queen’s University Belfast, David Keir Building,

Stranmillis Road, Belfast, BT9 5AG, UK

*corresponding author: Tel: +44 2890 974627, Fax: +44 2890 976524

Email: p.robertson@qub.ac.uk (P.K.J. Robertson)

2

Abstract

Microcystins and nodularin are toxic cyanobacterial secondary metabolites produced

by cyanobacteria that pose a threat to human health in drinking water. Conventional

water treatment methods often fail to remove these toxins. Advanced oxidation

processes such as TiO2 photocatalysis have been shown to effectively degrade

these compounds. A particular issue that has limited the widespread application of

TiO2 photocatalysis for water treatment has been the separation of the

nanoparticulate power from the treated water. A novel catalyst format, TiO2 coated

hollow glass spheres (Photospheres™), is far more easily separated from treated

water due to its buoyancy. This paper reports the photocatalytic degradation of

eleven microcystin variants and nodularin in water using Photospheres™. It was

found that the Photospheres™ successfully decomposed all compounds in 5

minutes or less. This was found to be comparable to the rate of degradation

observed using a Degussa P25 material, which has been previously reported to be

the most efficient TiO2 for photocatalytic degradation of microcystins in water.

Furthermore, it was observed that the degree of initial catalyst adsorption of the

cyanotoxins depended on the amino acid in the variable positions of the microcystin

molecule. The fastest degradation (2 minutes) was observed for the hydrophobic

variants (microcystin-LY, -LW, -LF). Suitability of UV-LEDs as an alternative low

energy light source was also evaluated.

Keywords: Cyanotoxins, Blue-green algae, UV-LEDs, Photospheres™, water

treatment

3

1. Introduction

Cyanobacteria and their secondary metabolites, in particular their toxins, pose a

serious health hazard to both humans and animals. Failure to control the

propagation of cyanobacteria can result in the production of potent toxic secondary

metabolites in aquatic systems. The management of cyanotoxins is important when

considering water safety, especially for potable use. Consequently the removal of

these toxins during water treatment is a key concern. The photocatalytic destruction

of cyanotoxins using TiO2, particularly microcystin-LR, has been studied in detail and

reported to be a very effective process for removal of these toxins from water [1-9].

The implementation of this technology by the water treatment sector, however, has

been limited due to the difficulties encountered in overcoming post-treatment catalyst

removal [10]. Three main ways of deploying TiO2 as a semi-conductor photocatalyst

in water treatment have been examined. The materials have been deployed as

(nanoparticulate) powders, as pellets, or as films attached to appropriate substrates

[11-13]. Powders generally perform well in the photocatalytic degradation of organic

compounds due to their large surface area and effective distribution in suspension.

Nevertheless, a significant challenge faced when using powders is the separation of

the powder suspension from the treated water [11]. Pelletised photocatalysts may

address the issue of catalyst separation, however, such materials tend to perform

less efficiently in comparison to the powders [10]. Another disadvantage of pelleted

photocatalysts is the fact that vigorous mixing can cause the pellets to fragment,

releasing free catalyst powder which results in the same issue of removal that is

encountered when using powders [10]. Photocatalysts can also be deployed as a

fixed matrix, usually as a film or in the form of nano tubes/rods. Preparation of these

materials, however often requires convoluted and cost-intensive manufacturing

4

processes, and they also have a reduced reactive surface area compared to powder

and pelletised materials [11,14]. A novel approach for deploying photocatalysts is

annealing the materials to hollow, buoyant glass spheres. The size of these products

ranges from the micro to millimetre. One of these novel products, known as

Photospheres™ (Table 1), are hollow buoyant TiO2 coated glass beads, which were

developed by Nanoparticulate Surface Adhesion Ltd. (NSA Ltd.). Photospheres™

are hollow glass beads (40µM diameter) coated with 100 % anatase TiO2 [15].

Insert Table 1 here

Due to the buoyancy of these materials they can be easily separated from treated

water, however, they still provide a high surface area. A number of researchers have

previously reported the use of TiO2 coated spheres or spheres formed from titania

for decomposition of contaminants in water [16-21]. Li et al. [16, 17] successfully

degraded methylene blue and orange II with titanium dioxide covered hollow silica

spheres. Zhao et al. [18] successfully degraded rhodamine B under visible light with

hollow spheres (Si/Ti hybrid) of Ag doped titania. Ren et al. [19] produced hollow

mesoporous TiO2 coated microspheres, however, they did not assess their

photocatalytic efficiency. There have also been two studies which have reported the

use of titania covered silica spheres of larger diameter (mm) for the control of

algae/cyanobacteria [20, 21]. These studies successfully targeted the growth of the

organisms but did not explore the removal of any of their harmful metabolites.

Photospheres™ combine the high surface area of particulate catalysts with

buoyancy. Jiang evaluated Photospheres™ for the photocatalytic destruction of

dimethyl phthalate and reported an optimum photocatalyst loading and irradiation

5

time of 8 g/L and 20 minutes respectively [22]. To date, no data has been published

examining the photocatalytic removal of microcystins using TiO2 coated

microspheres. This study is the first report of the application of Photospheres™ in

the photocatalytic degradation of eleven microcystin (MC) variants and nodularin in

water. While MC-LR is widely held to be the most commonly occurring microcystin, it

rarely occurs alone and to date many variants have been described [23]. It is

therefore important that a wide range of microcystin variants are evaluated to ensure

that degradation kinetics are comparable across a range of microcystin structures.

2. Experimental

2.1 Materials

Photospheres™ were purchased from NSA Ltd., Loanhead, UK. Microcystin variants

and nodularin were obtained as per Edwards et al. [24] (Enzo Life Sciences,

Farmingdale, USA). In addition to nodularin the microcystin variants studied included

microcystins -LR, -RR, -LA, -YR, -LY, -LW, -LF, HtyR, methylated microcystin-LR

and demethylated microcystins -LR and -RR. HPLC solvents were acetonitrile

(Rathburn, Walkerburn, UK) and Milli-Q water (Millipore, Watford, UK), trifluoroacetic

acid (TFA) was obtained from Fischer Scientific, Leicestershire, UK. All aqueous

solutions were prepared with Milli-Q water.

2.2 Photocatalysis

The photocatalytic method was derived from that reported by Robertson et al. [25]. A

13 mm, 4 mL screw top vial (Kinesis, Beds, UK) with a plastic lid with a self-healing

rubber septum and silicon facing was filled with a 10 µg mL-1 solution (3 mL) of a

6

microcystin variant or nodularin. One significant difference from previous protocols

was the application of mixing by air sparging as opposed to mechanical stirring as

previously described. Initial tests found that gentle stirring was not sufficient to mix

the catalyst throughout the reaction vessel due to its tendency to float. More vigorous

mixing, however, caused physical damage to the spheres as described by Mozia et

al. [26]. The test solution was air sparged by inserting a hypodermic needle through

the rubber septum of the vial and into the solution. Air flow was controlled via silicon

tubing connected to a rotameter (Influx Measurements, Alresford, UK), which, in turn,

was connected to an air pump (JUN-AIR, Nørresundby, Denmark; figure 1). The

airflow was maintained at 0.2 cm3 min-1. The reaction vessel was placed in front of a

Xenon lamp (480 W UVASpot 400 lamp, Dr Hönle UK, spectral output 330-450 nm,

light irradiance: 1230 µmol s-1 m-2) at a distance of 20 cm. One hundred twenty µL

samples were removed at the beginning of the experiment (T01) prior to the addition

of the photocatalyst allowing the initial concentration of analyte to be determined.

Subsequently the catalyst (1 % w/v TiO2; equivalent to 5.88 g Photospheres™ per

100 ml, suspension pH 5.12) was added to reaction vessel, mixed, then kept in the

dark for a further two minutes, after which another sample was taken (T02, indicative

of dark adsorption). The vial was subsequently exposed to the UV light source and

samples were taken at 1, 2, 3, 4, 5, 6, 8, and 10 minutes. Prior to analysis catalyst

was removed from the solution by small volume centrifugal filtration (5 minutes (2000

x g) at room temperature in Spin-X filters, 1.5 ml (Corning B.V. Life Sciences,

Amsterdam, The Netherlands). Two controls were performed: one in the dark to

confirm that decrease in concentration was due to photocatalytic activity alone, and

one without the catalysts present to assess the possible effect of UV irradiation alone

on the target analytes. All treatments were performed in duplicate. The photocatalytic

7

degradation of each microcystin and nodularin was determined individually with an

initial concentration of 10 µg mL-1 in water. To compare the toxin destruction on

Photospheres with that on the Degussa P25 material (Evonik Industries AG, Essen,

Germany) the degradation of MC-LR was performed with 1% w/v of the

photocatalyst with the pH of the P25 slurry at 5.33.

Insert figure 1 here.

2.3 Photosphere™ catalyst load and re-use

The effect of different catalyst loads on the photocatalytic degradation of microcystin-

LR was assessed. The experiments were carried out as described (section 2.2)

using a catalyst loading of between 0.2 % (w/v) TiO2 and 1 % (w/v) TiO2 in 0.2 %

increments.

Photosphere™ re-use for the degradation of MC-LR was evaluated using the same

initial set-up, however, sampling was only performed at 5 minutes. After sampling the

reactor was irradiated for a further 5 minutes to deplete the remaining MC-LR then a

new aliquot of MC-LR (10 µg mL-1) was added to the catalyst suspension. This was

repeated ten times.

2.4 Performance of Photospheres™ under UV-LED illumination

The recent increased availability of low energy UV-LEDs has prompted a number of

investigations into their application in photocatalytic waste remediation [26-30]. A

small scale reactor was designed to illuminate a 4 mL (13 mm diameter) glass screw

top vial centred in a ring of 30 LEDs. The reactor (PVC tube 51 x 45 mm, 4 mm wall)

was constructed by inserting LEDs in pre-drilled holes configured in three rows. The

UV-LEDs (AT Technologies, Bath, UK) had a diameter of 5 mm, a 15° aperture, λ

8

360 nm, and a power output of 750 μW at 20 mA DC /3.8 V. Each chain of 10 LEDs

was connected in series. The three LED chains were in turn connected in parallel.

The distance between the wall of the screw-top vial and the LEDs was 1.5 mm. The

degradation of MC-LR under UV-LED was evaluated for both Photospheres™ and

Degussa P25 with all other aspects of the photocatalysis as described (section 2.2)

2.5 Analysis

Analysis was performed by HPLC using a Waters 2695 Separation Module. High

resolution photodiode detection was performed with a Waters 2996 Photodiode

Array Detector (PDA) (both Waters, Elstree, UK). Separation of analytes was

performed with a Sunfire C18 column 2.1 mm (inner diameter) x 150 mm, with a 5

µm particle size (Waters, Elstree, UK). The mobile phases used were Milli-Q and

acetonitrile, both contained 0.05 % trifluoroacetic acid (TFA). Chromatography was

achieved over a linear gradient from 15% to 65% acetonitrile for 10 minutes followed

by a 100% solvent wash and equilibration. The flow rate applied was 0.3 mL min-1.

The PDA resolution was set to 1.2 nm and data was acquired in the range of 200 to

400 nm. Column temperature was set to 40°C [25].

3. Results and Discussion

3.1 Photocatalysis

Eleven microcystin variants and nodularin were irradiated in the presence of the

titanium dioxide coated Photospheres™. Results indicated that none of the

microcystin variants or nodularin were detected after five minutes continuous UV

irradiation time (Table 2).

9

Insert Table 2 here

Microcystin-LW, -LF, and -LY degraded fastest, with no detectable toxin remaining

after two minutes irradiation. Nodularin and Microcystin-HtyR degraded more slowly

with no detectable toxin remaining after four and five minutes respectively. There

were differing levels of the extent of dark adsorption to the photocatalyst surface for

the various microcystin variants with the rate of toxin degradation varying between

two and five minutes. Lawton et al. [32], previously reported that different microcystin

variants displayed different levels of dark adsorption on nanoparticulate Degussa

P25 powders. Similar trends appeared to follow for the microspheres examined in

this investigation. The results clearly showed that the amount of dark adsorption of

microcystins to TiO2 was dependent on the variable amino acid in the various

microcystin structures. The more hydrophobic microcystins, which contained leucine

at the variable position 4 and hydrophobic amino acids in position 2 (e.g.

microcystin-LW and -LF), tended to show a greater level of adsorption compared to

microcystin-RR which contained the more polar arginine at both variable positions.

This was also observed in Lawton et al.’s previous study on P25 TiO2 powders with

its conclusion that the hydrophobicity of the target analyte also played a role in dark

adsorption to the photocatalyst [33]. They also reported that the more hydrophobic

microcystin variants (microcystin-LW and microcystin-LF) had a higher level of dark

adsorption (at pH 4) than the less hydrophobic variant microcystin-RR. While the

overall extent of dark adsorption in this study differed in direct comparison, the

general relationship remained the same with microcystin-RR having the lowest dark

adsorption and microcystin-LW the highest. The comparative difference in the

amount of dark adsorption was most likely due to the properties of different catalysts

used. Degussa P25 has a BET surface area of approximately 50 m2 g-1 [33],

10

whereas Photospheres™ only have a BET surface area of approximately 27 m2 g-1

[15]. Another factor that could influence dark adsorption might be steric hindrance

posed by one, both, and/or the combination of the variable amino acids. Nodularin,

the smaller pentapeptide cyanotoxin was found to demonstrate the lowest dark

adsorption. Liu et al. [34] previously reported successful photocatalytic

decomposition of nodularin with TiO2 (Degussa P25), however, dark adsorption

differed greatly between that study and the present investigation (44 % compared to

14 %). Nonetheless, in the present study nodularin was undetectable within 4

minutes. These differences can be explained by the different components used in

the two studies (e.g. glass vessel thickness, mode of agitation, distance to light, toxin

concentration, and catalyst concentration).

Feitz et al. [35] and Lawton et al. [32] both reported a clear correlation between the

amount of dark adsorption and microcystin decomposition with the most effective

toxin removal being achieved where there was the greatest level of dark adsorption.

This observation, however, conflicts with a number of the microcystin variants

examined in this study. Microcystin-YR and methylated microcystin-LR displayed a

very similar level of dark adsorption (45 and 46 % respectively). When compared to

microcystin-LY (48 %), microcystin-YR and methylated microcystin-LR take twice as

long to degrade to a level where it was no longer detected. Furthermore, microcystin-

RR degrades almost twice as fast as Microcystin-HtyR, while having about half that

variant’s dark adsorption, although as stated above the overall difference in

degradation times between the different variants is relatively marginal.

Further studies exploring the catalytic degradation of different microcystin congeners

have recently been reported by He et al. [36] who examined the influence of variable

amino acids on the rate of degradation and mechanism of the destruction on four

11

microcystin variants, -LR, -YR, -RR and -LA, by both direct UV photolysis and three

advanced oxidation processes, including UV/H2O2, UVS2O82- and UV/HSO5-. They

reported that the variable amino acid not only influenced the overall rate of

degradation but also the reaction mechanism.

The degradation of microcystin-LR in the presence of 40 micron Photospheres™ and

Degussa P25 powder photocatalyst was compared (Figure 2). Encouragingly it can

be clearly seen that the degradation of the cyanotoxin using the Photospheres™ is

comparable with that achieved on the powder photocatalyst with the toxin being

decomposed within four minutes for both materials.

Insert Fig 2 here.

The decomposition of the toxin on both materials was also achieved under UV/LED

irradiation (Fig 3). Due to the lower photonic output of the LED arrays used in this

investigation the decomposition rates of microcystin-LR using both P25 and 40

micron Photospheres™ were slower than that observed under the xenon source. In

the case of P25 complete destruction of the toxin was achieved within 10 minutes

under UV LED irradiation, compared to four minutes under the xenon lamp. Using

the Photospheres™, however, over 30% of the microcystin remained after 10

minutes UVLED irradiation. The slower kinetics under UV LEDs is not surprising as

the influence of light intensity on the photocatalytic process has long since been

established [37]. The Photospheres™ appear to have been more significantly

influenced by the reduction in light intensity under the LED irradiation compared to

the P25 powders which may reflect the lower surface area of the Photospheres™.

Despite the reduction in removal efficiency the potential cost saving benefits of UV

LED illumination are worth considering with UV LED lamp life estimated at c.

12

100,000 hours compared to typical 1,000 hours for UV gas discharge sources.

Furthermore, the energy demand of the Xenon lamp used in this study was 450

Watts compared to the UV LED array with a radiant power of 12.05 mW. These

findings suggest significant efficiency gains can be made in designing treatment

systems based on LED technology.

Insert Fig 3.

3.2 Photosphere™ catalyst load and re-use

A series of catalyst loadings (0.2, 0.4, 0.6, 0.8, and 1.0 % TiO2) were tested. It was

found that while photocatalytic decomposition improved for loadings between 0.2 to

0.4, and 0.4 to 0.6 % (w/v) TiO2, only modest increases in photocatalytic efficiency

were obtained for catalyst concentrations of 0.6 % (w/v) TiO2 and above (Fig 4a).

This suggests that catalyst loadings of less than 1 % could achieve the desired

degree of toxin removal, in water treatment systems, reducing the overall cost of the

process. The relationship between catalyst loading and photocatalytic efficiency for

powder photocatalysts is well established [38, 39], with the optimal loading ranging

reported between 0.1 and 0.5 g L-1. Jiang et al., [22] reported optimum performance

of Photospheres™ at a dose of 8 g/L (0.8 %), and while the compounds being

treated were different, this is similar to the findings reported here. A linear

relationship between catalyst loading and the amount of dark adsorption was

observed (Fig 4b) for the range of catalyst loadings examined which would be

anticipated.

Insert Figure 4 here

13

The repeated use of the Photospheres™ for the degradation of microcystin-LR has

also been successfully demonstrated, with the photocatalytic efficiency for toxin

decomposition being maintained for up to ten cycles. This suggests that the

Photospheres™ could be recycled for continuous use in a water treatment process,

while maintaining their photocatalytic efficiency. Mozia et al. [26], however,

previously reported that over prolonged use in an aerated batch reactor the

Photospheres™ started to degrade and lose their buoyancy. In this study the

Photospheres™ appeared to maintain their integrity, buoyancy and photocatalytic

activity between repeated tests, however, treatment conditions may have to be

carefully designed to minimise damage.

4. Conclusion

It has been demonstrated that TiO2 coated silica spheres (Photospheres™) can

effectively decompose a range of microcystin variants and nodularin under UV

radiation. The rate of reaction was comparable to that achieved over a P25

photocatalyst for microcystin-LR. All the microcystin variants and nodularin were

degraded within less than five minutes concentrations above which these toxins are

usually detected in the environment. Degradation of microcystin-LR was also

achieved under UV LED irradiation for both P25 and Photospheres™, although the

rate of decomposition was significantly lower due to the lower photonic output of the

UV LEDs compared to the Xenon source. This suggests that UV photocatalysis with

Photospheres™ could be a viable method for the treatment of water contaminated

with microcystin and/or nodularin. Further research is required to determine the

viability of the application in more complex matrices and at more environmentally

relevant concentrations.

14

Acknowledgements

We thank the European Cooperation in Science and Technology COST Action

ES1105 CYANOCOST for networking and knowledge-transfer support. The authors

would also like to thank Dr J.M.C. Robertson and Mr L.J. Montgomery for their

support in the preparation of this manuscript. The authors also thank Dr Jay Murphy

for proof reading the manuscript.

5. References

1. L.A. Lawton, P.K.J. Robertson, Chemical Society Reviews, 28 (1999) 217-224

2. B.J.P.A. Cornish, L.A. Lawton, P.K.J. Robertson, Applied Catalysis B:

Environmental 25 (2000) 59-67

3. I. Liu, L.A. Lawton, B.J.P.A. Cornish, P.K.J. Robertson, Journal of

Photochemistry and Photobiology A: Chemistry, 148 (2002) 349-354

4. G.S. Shephard, S. Stockenstrom, D. de Villiers, W.J. Engelbrecht, G.F.S.

Wesels, Water Research 36 (1) (2002)140-146

5. X.G. Fieng, F. Rong, D.G. Fu, C.W. Yuan, Y. Hu, Chinese Science Bulletin, 51

(2006) 1191-1198

6. M.G. Antoniou, P.A. Nicolaou, J.A. Shoemaker, A.A de la Cruz and D.D.

Dionysiou, App. Catal. B: Environ., 91 (2009) 165-173.

7. M. G. Antoniou, J. A. Shoemaker, A. A. de la Cruz, D. D. Dionysiou, Environ. Sci.

Technol. 42, (2008) 8877–8883

8. M.G. Antoniou, J.A. Shoemaker, A.A de la Cruz and D.D. Dionysiou, Toxicon, 51

(2008) 1103-1118.

15

9. H. Choi, M.G. Antoniou, M. Pelaez, A.A. De la Cruz, J.A., Shoemaker, D.D.

Dionysiou, Environ. Sci. Technol. 41, (2007) 7530-7535.

10. I. Liu, L.A. Lawton, D.W. Bahnemann, L. Liu, B. Proft, P.K.J. Robertson,

Chemosphere 76 (2009) 549-553

11. S. Meyer, R. Gorges, G. Kreisel, Thin Solid Films, 450:2 (2004) 276-281

12. R.S. Sonawane, S.G. Hegde, M.K. Dongare, Materials Chemistry and Physics,

77 (2002) 744-750

13. H. Imai, M. Matsuta, K. Shimizu, H. Hirashima, N. Negishi, Journal of Materials

Chemistry 10 (2000) 2005-2006

14. S. Kobayashi, K. Hanabusa, N. Hamasaki, M. Kimura, H. Shirai, Chemistry of

Materials 12:6 (2000) 1523-1525

15. M.R. Banks, Process for the manufacture of titania coated microspheres (2011)

United States Patent Application No: 20110052804

16. G. Li, F. Liu, Z. Zhang, Journal of Alloys and Compounds 493:1-2 (2010) L1-L7

17. G. Li, R. Bai, X.S. Zhao, Industrial Engineering Chemistry Research 47:21 (2008)

8228-8232

18. W. Zhao, L. Feng, R. Yang, J. Zheng, X. Li, Applied Catalysis B: Environmental,

103:1-2 (2011) 181-189

19. T-Z. Ren, Z-Y, Yuan, B-L, Su, Chemical Physics Letter, 374:1-2 (2003) 170-175

20. S-C. Kim, D-K Lee, Microchemical Journal, 80 (2005) 227-232

21. J.R. Peller, R.L. Whitman, S. Griffith, P. Harris, C. Peller, J. Scalzitti, Journal of

Photochemistry and Photobiology A: Chemistry 186 (2007) 212-217

22. W. Jiang, J. A. Joens, D. D. Dionysiou, K. E. O’Shea, Journal of Photochemistry

and Photobiology A:Chemistry, 262 (2013) 7-13

16

23. L.A. Lawton and C. Edwards, Purification of Microcystins. Journal of

Chromatography A, 912 (2001) 191-209.

24. C. Edwards, L.A. Lawton, S.M. Coyle, Journal of Chromatography A 734 (1996)

163-173

25. P.K.J. Robertson, L.A. Lawton, B. Munch, J. Rouzade, Chemical

Communications, 4 (1997) 393-394

26. S. Mozia, P. Brozek, J. Przepiórski, B. Tryba, M. Morawski, J. Nanomat., 2012,

Article ID 949764

27. O.I. Tokode, R. Prabhu, L.A Lawton and P.K.J. Robertson, Journal of Catalysis,

(2012), 290, 138-142.

28. Y. Sun, W. Ling Zhang, S. Sun, Materials Research Bulletin, 48 (2013) 4357-

4361

29. E. Korovin, D. Selishchev, A. Besov, D. Kozlov, Applied Catal. B: Environ., 163

(2015) 143-149

30. I. Levchuk, J.J. Rueda-Márquez, S. Suihkonen, M.A. Manzano, M. Sillanpää,

Separation and Purification Technology, 143 (2015) Pages 1-5

31. O.I. Tokode, R. Prabhu, L.A Lawton and P.K J. Robertson, “UV LED sources for

heterogeneous photocatalysis” in “The Handbook of Environmental Chemistry,

Vol. 35 Environmental Photochemistry Part III”, eds. Detlef W. Bahnemann

and Peter K.J. Robertson, Springer-Verlag Berlin Heidelberg. (2015), in press.

32. L.A. Lawton, C. Edwards, G. Codd, Analyst 119 (1994) 1525-1530

33. L.A. Lawton, P.K.J. Robertson, B.J.P.A. Cornish, I.L. Marr, I.L., M. Jaspars,

Journal of Catalysis 213 (2003) 109-113

34. I. Liu, L.A. Lawton, D.W. Bahnemann, P.K.J. Robertson, Applied Catalysis B:

Environmental 60 (2005) 245-252

17

35. A.J. Feitz, T.D. Waite, G.J. Jones, B.H. Boyden, P.T. Orr, Environmental Science

and Technology 33 (1999) 243-249

36. X. He, A. de la Cruz, A. Hiskia, T. Kaloudis, K. O’Shea, D. Dionysiou, Water

Research, 74 (2015) 227-238

37. M.R. Hoffmann, S.T. Martin, W Choi and D.W. Bahnemann, Chem. Rev. 95

(1995) 69-96.

38. R. Andreozzi, V. Caprio, A. Insola, G. Longo, V. Tufano, Journal of Technology

and Biotechnology, 75:2 (2000) 131-136

39. P.A. Pekakis, N.P. Xekoukoulotakis, D. Mantzavinos, Water Research, 40:6

(2006) 1276-1286

18

List of Captions for Figures.

Fig. 1 Photocatalytic reactor for Toxin Destruction using Photospheres™

Fig. 2 Photocatalytic degradation of microcystin-LR in the presence of 1% (w/v TiO2) Photospheres™(); Degussa P25 ().Error bars=1 SD; n=2.

Fig. 3 Degradation of microcystin-LR using UV-LEDs as source of radiation in the presence of 1% (w/v TiO2) Photospheres™(); Degussa P25 (). Error bars=1 SD; n=2.

Fig. 4 (A) The effect of different catalyst loads (0.2 %(); 0.4 % (); 0.6 %(); 0.8 % ();1.0 % TiO2 () of Photospheres™ on the photocatalytic decomposition of microcystin-LR. (B) Dark adsorption of microcystin-LR to Photospheres™ at different catalyst loads. Error bars=1 SD; n=2.

Table 1 Properties and SEM observation of Degussa P25 (bar = 1 μm) and Photospheres™ (bar = 10 μm) [data from references 10, 15 and the present study]

Table 2 Summary of the degradation and dark adsorption of 11 different microcystin variants and nodularin with Photospheres™.

19

Fig 1.

Catalyst in solution (3 mL)

4 mL glass vial

screw lid -loosely closed

hypodermic needle

silicone tubing

rotameterair pump

20 cm

UV light

20

Fig 2.

0123456789

10

0 2 4 6 8

Co

nce

ntr

atio

n (μ

g m

L-1)

Time (min)

UV light on

21

Fig. 3

0

2

4

6

8

10

12

0 2 4 6 8 10 12

Conc

entr

atio

n M

C-LR

(μg

L-1)

Time (min)

UV light on

22

Fig 4.

A

0123456789

10

0 2 4 6 8

Co

nce

ntr

atio

n (μ

g m

L-1)

Time (min)

UV light on

R² = 0.9978

30

25

20

15

5

10

B

Dar

k A

dso

rpti

on M

C-L

R (

%)

0 0 0.2 0.4 0.6 0.8 1

Concentration TiO2 (%)

23

Material Properties Image Degussa P25 Nanoparticulate powder 100% TiO2

BET surface area: 50 m2 g-1 Particle size approx 25 nm TiO2 composition: 75% anatase

Photospheres™ Coated silica beads 17% TiO2

BET surface area: 27 m2 g-1 Particle size 40 µm (10 - 60 µm) TiO2 Composition: 100% anatase

Table 1.

* Time at which no microcystin/nodularin could be detected by HPLC

Table 2.

Microcystin Variant Dark adsorption (%) Complete degradation* (min)

Nodularin 14 4

Microcystin-RR 21 3

Microcystin-LR 27 5

Demethylated Microcystin-RR 32 4

Microcystin-LA 34 4

Microcystin-HtyR 43 5

Demethylated Microcystin-LR 44 5

Methylated Microcystin-LR 45 4

Microcystin-YR 46 4

Microcystin-LY 48 2

Microcystin-LW 64 2

Microcystin-LF 70 2

24